{"id":"047878b9-de25-417f-85d2-63fd2be74968","arxiv_id":"1908.08253","paper_version":2,"verdict":"CONDITIONAL","confidence":"MODERATE","novelty_score":6.0,"correctness_risk":"high","formal_verification":"none","parameter_count":0,"one_line_summary":"For every d at least 2, the full Kontsevich graph complex maps injectively into the Chevalley-Eilenberg complex of the n=d-1 Schouten algebra, so its zeroth cohomology acts via L-infinity automorphisms on graded symplectic manifolds.","lead":"This paper builds a graph-complex model for the deformation theory of graded symplectic manifolds of any degree, extending Kontsevich's original construction for Poisson manifolds. General readers may care because it yields explicit new universal deformation formulas, including flows on Courant algebroids.","discovery_kind":"new_application","skeptic_critique":{"model":"deepseek-v4-flash","headline":"Local sign inconsistency in Rep(d): under the paper's literal formulas the single-edge graph is not the graded Poisson bracket, so the central operad morphism fails before any globalization issue is reached.","rationale":"The reader's weakest assumption was that the local Darboux formulas do not patch to a globally well-defined action. I agree that global well-definedness is unproved, but I find a more immediate and more damaging problem: the explicit bilinear operator Delta_{ij} in (5.10)-(5.11) does not equal the Poisson bracket of (3.4)-(3.9) under the paper's stated sign conventions. Since Proposition 5.2 is used to identify the single-edge graph with the bracket, and since the proof of Proposition 5.1 is only a 'can be checked' statement, the tower of representations and all subsequent universal-structure results rest on an unverified and, as written, false local identity. The paper does give credit-worthy standard material: the review of graph complex cohomology, the Willwacher results, and the explicit Courant algebroid flow equations are concrete and checkable. But those flow equations inherit the sign conventions of Rep(d), so they cannot be trusted until the sign issue is resolved. The concern is serious enough to prevent acceptance in the current form, yet it is plausibly fixable by correcting the operators and proving the equivariance with proper Koszul signs, so a conditional recommendation rather than outright rejection is appropriate. The reader's verdict of CONDITIONAL is therefore kept, but the stated reason should be broadened from a purely global-patching gap to the more fundamental local sign inconsistency in the representation.","tokens_in":42954,"tokens_out":24224,"duration_ms":264566,"concrete_test":"Evaluate Proposition 5.2 in the simplest nontrivial case: V=T*[1]R with Darboux coordinates (x,p), p odd of degree 1, f=α(x)p, g=β(x)p. Using the manuscript's definitions, compute Rep_2^{(2)}(Γ)(f⊗g) from (5.10) and compare it with {f,g} from (3.8). The two expressions are p(α'β+αβ') and p(αβ'-α'β) respectively; any nonzero difference disproves the asserted equality Rep_2^{(2)}(Γ)={·,·}, hence the operad morphism property underlying (1.4). An independent re-derivation of (5.8) including Koszul signs, followed by the same one-dimensional check, would show whether the intended construction is repairable.","verdict_should_be":"CONDITIONAL","load_bearing_attack":"The central construction is the operad morphism Rep(d): Grad -> End_{C^∞(V)} of Proposition 5.1, which is then used in Proposition 5.3 to obtain the injective dg Lie algebra morphism fGC_d -> CE(T_poly^{(n)}). If Rep(d) is not a morphism of operads, the main claim (1.4) and Corollary 5.6 do not follow. As written, the d=2 case of (5.10) gives Delta(f,g)=∂_x f ∂_p g + ∂_p f ∂_x g, while the Poisson bracket in (3.8) has a degree-dependent sign: {f,g}=(-1)^{deg f} ∂_x f ∂_p g + ∂_p f ∂_x g. These differ already on degree-1 functions on T*[1]R: for f=α(x)p and g=β(x)p, the bracket is p(αβ'-α'β), whereas (5.10) gives p(α'β+αβ'). Moreover, the S_N-action on the endomorphism operad stated in (5.8) omits the Koszul signs that would reconcile the graded antisymmetry of the bracket with the symmetrization of the graph Γ. The proof of Proposition 5.1 merely states that properties 1-3 'can be checked' and does not address these signs. The global coordinate-independence issue raised in the reader report is real, but it is secondary: the local formulas are not even consistent with Proposition 5.2 under the paper's own sign conventions.","agreement_with_reader":"partial"},"referee_report":{"model":"deepseek-v4-flash","summary":"The paper proposes a generalization of Kontsevich's graph-complex construction from polyvector fields (n = 1) to graded symplectic manifolds of arbitrary degree n. The central object is a claimed tower of operad morphisms Rep(d): Grad -> End_{C^∞(V)}, where V is any NP-manifold of degree n and d = n + 1, defined by explicit local Darboux-coordinate formulas. From this the paper derives an injective morphism of dg Lie algebras fGC_d -> CE(T_poly^(n)), an action of exp(H^0(fGC_d)) by Lie-infinity automorphisms of the n-Schouten algebra, a classification of universal structures, universal Hamiltonian deformations and flows, and applications to Courant algebroids. The paper also proposes explicit new flow equations for Courant algebroids and conformal Hamiltonian flows from trivalent graph cohomology.","tokens_in":43199,"tokens_out":10724,"duration_ms":102702,"significance":"If the central representation were correct and globally well defined, the paper would provide a valuable unifying framework extending the Grothendieck-Teichmueller action on polyvector fields to higher graded symplectic manifolds, with concrete and checkable formulas for Courant algebroids. The explicit expressions (6.11)-(6.12) and the conformal factors (6.13)-(6.14) are useful and nontrivial. The paper draws on substantial external results on graph cohomology and does not provide machine-checked proofs; its value depends on the correctness of the local representation and on the completeness of the cohomological classification. As written, the central claim is not established because of the sign inconsistencies, missing globalization argument, and unsupported classification lemma discussed below.","major_comments":[{"comment":"The local sign conventions are inconsistent with the Poisson bracket (3.8). For d = 2, eq. (5.10) gives Delta_12(f⊗g) = ∂_x f ∂_p g + ∂_p f ∂_x g, whereas the bracket in (3.8) is {f,g}_ω = (-1)^{deg f} ∂_x f ∂_p g + ∂_p f ∂_x g. On T*[1]R, with f = α(x)p and g = β(x)p, both of degree 1, one obtains {f,g}_ω = p(αβ' - α'β), while Delta_12(f⊗g) = p(α'β + αβ'). These are different, so eq. (5.13) fails and Rep(d) is not the asserted operad morphism; consequently Proposition 5.3, the injective morphism (1.4), and Corollary 5.6 do not follow as written. The proof of Proposition 5.1 only states that properties 1-3 'can be checked' and does not address the Koszul signs that would be needed in the graded S_N-action (5.8). A similar problem occurs for odd d with the ξ-term, which cancels under the orientation morphism when κ is symmetric.","section":"Section 5, eqs. (5.10), (5.13) and Proposition 5.2"},{"comment":"The representation is defined by local Darboux-coordinate formulas (5.5)-(5.11), but no proof is given that these operators are independent of the chosen graded Darboux chart or that they define global differential operators on an arbitrary NP-manifold. The claimed morphism Rep(d): Grad -> End_{C^∞(V)} and the induced morphism (5.15) are asserted for every NP-manifold, and all subsequent classification and action statements depend on this global well-definedness. For the case n = 1, the literature shows that a globalization argument is needed (see [66]); the paper does not provide the analogous argument for n > 1.","section":"Section 5, Proposition 5.1"},{"comment":"Lemma 5.7 states a complete low-degree cohomology classification of fGC_d^con for all d ≠ 2, but Section 4.3 only reports cohomological information for d = 2 and d = 3. The 'various bounds collected in Section 4.3' do not include the vanishing statements for H^0, H^1, and H^2 of GC_d for general d that the lemma requires. Consequently, Proposition 5.8's uniqueness claims — the unique universal automorphism for n = 4j+2 and the unique universal deformation for n = 4j+3 — are not supported by the cited material. The authors should either prove Lemma 5.7 or provide precise references for every vanishing result entering the classification.","section":"Section 5, Lemma 5.7"}],"minor_comments":[{"comment":"The symbol Γ is used both for the edge graph in (4.8) and for the one-vertex graph with no edges in the proof of Proposition 5.13; the assertion 'δ Γ = Γ' is therefore confusing and should be clarified with distinct notation.","section":"Section 5, Proposition 5.13"},{"comment":"The proof of Proposition 5.2 consists of the single equality (5.13); given the sign issue in eq. (5.10), a full verification of the operad identities is needed rather than a one-line assertion.","section":"Section 5, Proposition 5.2"},{"comment":"The introduction and Section 6.2 speak of deformations of Courant algebroids, but the constructed flows deform the Courant-Dorfman structure while leaving the fiber metric fixed; this should be stated more prominently to avoid overstatement.","section":"Section 6.2"}],"recommendation":"major_revision","confidential_remarks":"This is an ambitious preprint with a companion paper in preparation. The central construction currently fails as written due to local sign inconsistencies and lacks a globalization proof; the classification lemma is also unsupported by the cited material. I recommend major revision rather than rejection because the overall framework is plausible and the explicit formulas may be repairable within the scope of a revised manuscript."},"author_rebuttal":null,"desk_editor":{"model":"deepseek-v4-flash","letter":"What you should know: this paper tries to extend Kontsevich's graph-complex machinery to graded symplectic manifolds of arbitrary degree, and it contains genuinely new explicit formulas, especially the Courant algebroid flows. But the central representation has a sign problem: the single-edge graph does not map to the graded Poisson bracket. That breaks the main morphism claim before the harder globalization question is even reached.\n\nWhat is good: the overall idea is the right one—using the full graph complex fGC_d on NP-manifolds of degree n=d−1 is natural, and the paper gives a clear review of the needed graded geometry and graph complex background. The explicit operators (5.10)-(5.11), the triangle flow (6.11)-(6.12), and the conformal flow construction are concrete and checkable. If the sign issue is repairable, the classification of universal structures on higher Courant algebroids would be a real contribution. The loop-induced actions and the conformal Hamiltonian flows are also novel and worth taking seriously.\n\nThe soft spots are serious. First, the local formulas are inconsistent with the paper's own bracket conventions. For d=2, the bracket in (3.8) is {f,g} = (-1)^{deg f} ∂_x f ∂_p g + ∂_p f ∂_x g, while (5.10) gives the symmetric expression without the sign. Testing on f=α(x)p and g=β(x)p gives p(αβ'−α'β) for the bracket but p(α'β+αβ') for the representation. The same issue appears in d=3 with the ξ-term. Since Proposition 5.3 and Corollary 5.6 hinge on s^*Rep(d)(Γ) = [·,·]_S, the main injective morphism (1.4) does not follow as written. This is not a minor typo; it is load-bearing. Second, the paper claims a complete low-degree cohomology classification in Lemma 5.7 for all d≠2, but the supporting bounds in Section 4.3 are only stated for d=2 and d=3. That lemma is unsupported. Third, the representation is defined in local Darboux coordinates and no argument shows it patches globally; this is secondary to the sign problem but still needs addressing.\n\nWho is this for? Anyone working on deformation theory of Courant algebroids or on universal graph complexes will want to know about the ideas here, but they should not rely on the formulas without checking the signs first. The paper deserves a serious referee—the flaws look fixable and the framework is valuable—but it needs major revision. I would recommend sending it to peer review and asking the authors to correct the sign conventions, prove coordinate independence, and either prove or weaken Lemma 5.7.","headline":"Interesting generalization with a load-bearing sign error: the central representation as written does not reproduce the Poisson bracket, so the main theorem is unproven.","tokens_in":43806,"tokens_out":12650,"would_cite":false,"duration_ms":120952,"reading_group":"maybe","serious_thinker":"no","would_accept_peer_review":true},"rs_alignment":null,"lean_confirmation":null,"pith_extraction":{"msc":["17B55","18D50","53D17","58A50"],"pacs":[],"model":"deepseek-v4-flash","headline":"This paper extends the injective graph-complex morphism from polyvector fields to every graded symplectic manifold, with $d=n+1$.","keywords":["graph complexes","graded symplectic manifolds","L-infinity automorphisms","Schouten algebra","Grothendieck-Teichmueller algebra","Courant algebroids","AKSZ sigma-models","universal deformations"],"falsifier":"Compute the image of the triangle graph under $\\mathrm{Rep}^{(3)}$ for one Hamiltonian on a degree-2 NP-manifold in two overlapping graded Darboux charts. If the two local expressions for $\\dot{\\rho}^a{}_\\mu$ and $\\dot{T}_{abc}$ in (6.11)–(6.12) do not coincide on the overlap, the operad representation is not global and the injection $\\mathsf{fGC}_3 \\hookrightarrow \\mathsf{CE}(T^{(2)}_{\\mathrm{poly}})$ fails on nontrivial manifolds.","tokens_in":42649,"feed_emoji":"🕸️","tokens_out":9308,"duration_ms":84357,"temperature":0.7,"pith_summary":"The paper claims that the full graph complex $\\mathsf{fGC}_d$ embeds injectively into the Chevalley–Eilenberg complex of the $n$-Schouten algebra $T^{(n)}_{\\mathrm{poly}}$ for every $d=n+1\\ge 2$, where $n$ is the degree of a graded symplectic (NP) manifold. If correct, this makes the cohomology of graph complexes a classification tool for universal structures on all graded symplectic manifolds, not only on Poisson manifolds. In particular, $\\exp(H^0(\\mathsf{fGC}_d))$ acts by $\\mathrm{Lie}_\\infty$-automorphisms on the $n$-Schouten algebra, generalizing the known action of the Grothendieck–Teichmüller Lie algebra on polyvector fields. The paper works out the case of Courant algebroids: a unique triangle-graph flow deforms Courant–Dorfman structures, and trivalent graphs modulo IHX produce conformal Hamiltonian flows.","feed_headline":"One graph complex now acts on every graded symplectic manifold","feed_subtitle":"The same graph machinery now governs all graded symplectic manifolds, from Poisson manifolds to Courant algebroids.","key_machinery":"The load-bearing object is the tower of representations $\\mathrm{Rep}^{(d)}:\\mathsf{Gra}_d \\hookrightarrow \\mathrm{End}_{C^\\infty(V)}$ that turns graphs into differential operators. Vertices are decorated by functions on the NP-manifold and each edge $(i,j)$ is read as the second-order operator $\\Delta_{ij}$ of (5.10) for $d$ even or (5.11) for $d$ odd; these operators carry degree $1-d$ and obey the edge-permutation and orientation-flip signs of the graph operad. Passing through the orientation morphism $\\mathsf{Gra}_d \\hookrightarrow d\\mathsf{Gra}_d$ and taking Chevalley–Eilenberg cochains, the same data produces the injection $\\mathsf{fGC}_d \\hookrightarrow \\mathsf{CE}(T^{(n)}_{\\mathrm{poly}})$, so cohomology classes in the graph complex become universal infinitesimal deformations and automorphisms.","core_discovery":"The central claim is an injective morphism of dg Lie algebras $\\mathsf{fGC}_d \\hookrightarrow \\mathsf{CE}(T^{(n)}_{\\mathrm{poly}})$ for every $d\\ge 2$, with $T^{(n)}_{\\mathrm{poly}} = C^\\infty(V)[n]$ the $n$-Schouten algebra of an NP-manifold of degree $n$ and $d=n+1$. The injection is induced by a tower of operad morphisms $\\mathrm{Rep}^{(d)}:\\mathsf{Gra}_d \\hookrightarrow \\mathrm{End}_{C^\\infty(V)}$, where each graph edge acts by the bidifferential operator $\\Delta_{ij}$ written in local graded Darboux coordinates. From this, the paper derives that $\\exp(H^0(\\mathsf{fGC}_d))$ acts via $\\mathrm{Lie}_\\infty$-automorphisms, that stable universal deformations of the $n$-Schouten algebra are generated by loop cocycles, and that for $d=3$ the loop $L_3$ gives the unique universal deformation flow for Courant algebroids, with explicit components displayed for the anchor and the three-form.","pith_inferences":["The paper states the representation formulas in local graded Darboux coordinates and does not prove that they glue to a global representation under changes of coordinates; if coordinate invariance fails, the classification would hold only locally or for a restricted class of NP-manifolds.","A natural testable extension is to compute the triangle flow (6.11)–(6.12) on a nontrivial Courant algebroid beyond the local split form, or to compare it with the perturbative expansion of the Courant $\\sigma$-model; a match would anchor the algebraic graph complex in AKSZ field theory.","The absence of $\\mathfrak{grt}_1$ in stable higher-dimensional structures suggests that any incarnation of the Grothendieck–Teichmüller algebra in Courant-type deformation theory has to come from oriented or multi-oriented graph complexes, as the paper announces for a companion paper.","If the one-dimensional $H^0(\\mathsf{fGC}_3)$ result is read as a rigidity statement, it implies that universal deformation theory for Courant algebroids is much more constrained than for Poisson manifolds, so new phenomena must be sought either in non-stable cochains or in additional geometric data such as the anchor and three-form."],"forward_implications":["For $d\\ne 2$, stable universal structures on graded symplectic manifolds come only from loop cocycles: degree $n=4j+2$ admits the unique $\\mathrm{Lie}_\\infty$-automorphism from $L_{4j+3}$, and degree $n=4j+3$ admits the unique formal deformation from $L_{4j+5}$.","In dimension $d=3$, the vanishing $H^1(\\mathsf{fGC}_3)=0$ removes stable obstructions to a universal formality morphism for Courant algebroids, while $H^0(\\mathsf{fGC}_3)=\\mathbb{K}$ predicts a one-parameter family of stable universal quantization maps.","The triangle cocycle $L_3$ produces an explicit universal flow on the space of Courant–Dorfman structures, given in components by equations (6.11)–(6.12), with deformed anchor and three-form.","Trivalent graphs modulo the IHX relation produce conformal Hamiltonian flows on Courant algebroids, with conformal factors such as $\\Omega_\\Theta = T_{abc}T^{abc}+6\\,\\partial_\\nu\\rho^a{}_\\lambda\\,\\partial^\\lambda\\rho_a{}^\\nu$.","For $d=1$ the same machinery reproduces the Groenewold–Moyal product, and for $d=2$ it recovers the Grothendieck–Teichmüller group action and the tetrahedral flow on Poisson bivectors."],"supporting_citations":[{"why":"introduces the original universal graph complex $\\mathsf{fGC}_2$ and its local morphism into the Chevalley–Eilenberg complex of polyvector fields, the construction this paper generalizes.","marker":"[78]"},{"why":"provides the formality morphism and admissible-graph calculus whose supergeometric interpretation underlies the representation of graphs on functions.","marker":"[79]"},{"why":"supplies the classification of graded symplectic manifolds and their local Darboux coordinates, including the bijection between degree-2 NPQ-manifolds and Courant algebroids.","marker":"[99]"},{"why":"defines the graph operad $\\mathsf{Gra}_d$, the full graph complex $\\mathsf{fGC}_d$, its cohomology, and the identification $H^0(\\mathsf{GC}_2)\\cong\\mathfrak{grt}_1$ that the paper extends.","marker":"[113]"},{"why":"establishes the stable formality setting and the regular action of $\\exp(H^0(\\mathsf{fGC}_2))$, the mechanism whose generalization is Corollary 5.6.","marker":"[32]"},{"why":"supplies cohomology computations for graph complexes, including loop classes and trivalent graphs modulo IHX, used to classify universal structures in low degrees.","marker":"[70]"}],"fun_headline_variants":["Graph complex tames all graded symplectic manifolds","Kontsevich graphs now deform Courant algebroids","From Poisson to Courant: one graph complex to rule them all","Universal deformations for all graded symplectic manifolds"],"cache_read_input_tokens":3200,"weakest_assumption_plain":"The load-bearing premise is that the local differential operators (5.10)–(5.11) piece together into a single well-defined representation of the graph operad on every NP-manifold, invariant under changes of graded Darboux coordinates; the paper gives local formulas but no global coordinate-invariance proof.","fun_headline_variants_meta":{"raw":{"variants":["Graph complex tames all graded symplectic manifolds","Kontsevich graphs now deform Courant algebroids","From Poisson to Courant: one graph complex to rule them all","Universal deformations for all graded symplectic manifolds"]},"model":"deepseek-v4-flash","effort":"low","cost_usd":0.000617,"raw_usage":{"total_tokens":2969,"prompt_tokens":1153,"completion_tokens":1816,"prompt_tokens_details":{"cached_tokens":384},"prompt_cache_hit_tokens":384,"prompt_cache_miss_tokens":769,"completion_tokens_details":{"reasoning_tokens":1745}},"tokens_in":769,"tokens_out":1816,"duration_ms":13345,"temperature":1.0,"reasoning_tokens":1745,"cache_read_input_tokens":384,"cache_creation_input_tokens":0},"cache_creation_input_tokens":0},"created_at":"2026-08-14T11:45:42.598100+00:00","model_set":{"reader":"deepseek-v4-flash"},"falsifier":"Compute the image of the triangle graph under $\\mathrm{Rep}^{(3)}$ for one Hamiltonian on a degree-2 NP-manifold in two overlapping graded Darboux charts. If the two local expressions for $\\dot{\\rho}^a{}_\\mu$ and $\\dot{T}_{abc}$ in (6.11)–(6.12) do not coincide on the overlap, the operad representation is not global and the injection $\\mathsf{fGC}_3 \\hookrightarrow \\mathsf{CE}(T^{(2)}_{\\mathrm{poly}})$ fails on nontrivial manifolds.","supporting_citations":[{"cited_title":"M. Kontsevich's graph complex and the Grothendieck-Teichmueller Lie algebra","cited_arxiv_id":"1009.1654","evidence_quote":"defines the graph operad $\\mathsf{Gra}_d$, the full graph complex $\\mathsf{fGC}_d$, its cohomology, and the identification $H^0(\\mathsf{GC}_2)\\cong\\mathfrak{grt}_1$ that the paper extends."},{"cited_title":"Stable Formality Quasi-isomorphisms for Hochschild Cochains","cited_arxiv_id":"1109.6031","evidence_quote":"establishes the stable formality setting and the regular action of $\\exp(H^0(\\mathsf{fGC}_2))$, the mechanism whose generalization is Corollary 5.6."},{"cited_title":"Differentials on graph complexes","cited_arxiv_id":"1411.2369","evidence_quote":"supplies cohomology computations for graph complexes, including loop classes and trivalent graphs modulo IHX, used to classify universal structures in low degrees."}],"review_version":1}